PaddleOCR Document Parsing

PaddleOCR Document Parsing

by PaddlePaddle
Apache-2.0
Repository
markdown Document Processing
ocrdocument-processingpdflayout-parsingdata-extraction

Summary

Turn PDFs and scanned images into structured Markdown/JSON: cell-level tables, LaTeX formulas, correct multi-column order.

Features

  • Structured Markdown/JSON output preserving tables, formulas, and reading order
  • Table extraction at cell-level precision, not just flat text
  • Math formulas recognized and output as LaTeX
  • Handles multi-column layouts with correct reading order

Install This Skill

Add this skill to your favorite AI agent in a few steps.

Any AI agent

This skill is plain instructions — it works with any assistant that accepts custom instructions or system prompts.

  1. Copy the skill content with the button below.
  2. Paste it into your agent's instruction file or system prompt (for example AGENTS.md, .cursorrules, or a custom instructions field).
  3. Ask the agent to apply the skill whenever the task matches.

Skill Content

Markdown Content

Copy this content and use it with your preferred AI agent

# PaddleOCR Document Parsing

## When to Use This Skill
Use this skill for:
- Documents with tables (invoices, financial reports, spreadsheets)
- Documents with mathematical formulas (academic papers, scientific documents)
- Documents with charts and diagrams
- Multi-column layouts (newspapers, magazines, brochures)
- Complex document structures requiring layout analysis

## Usage
### Basic Document Parsing
From URL:
```bash
paddleocr api --model_type doc_parsing --file_url "https://example.com/report.pdf"
```
From local file:
```bash
paddleocr api --model_type doc_parsing --file_path "./document.pdf"
```

Description

PaddleOCR Document Parsing is the official agent-skill wrapper around PP-StructureV3 and PaddleOCR-VL, the document-layout models from Baidu's widely used PaddleOCR project. Rather than treating a document as a flat block of text, it drives the paddleocr CLI's doc_parsing model to output structured Markdown or JSON that preserves what matters in complex documents: table cells with correct row and column structure, mathematical formulas as LaTeX, figures, seals and charts, headers and footers, and correct reading order across multi-column layouts such as newspapers, financial reports, and academic papers.

It accepts a local file path or a remote URL and needs only a PaddleOCR access token plus the paddleocr binary, which is auto-installable via uv. Because it targets structure rather than plain OCR text, it is the right tool when a document's value is in its layout, such as an invoice's line-item table, a financial report's charts, or an academic paper's equations, where a flat text dump would lose that information. Use it whenever a complex PDF or scanned image needs to become something an LLM or downstream pipeline can reason over.

Related Skills

Docling Document Intelligence
Featured

Docling Document Intelligence

by docling-project (IBM)

Parse, convert, chunk, and structurally analyze PDFs, DOCX, PPTX, HTML, and images with IBM's open-source Docling toolkit.

Document Processing
document-processingragpdf
GLM-OCR Table Recognition

GLM-OCR Table Recognition

by Z.ai (Zhipu AI)

Extract complex tables, including merged cells and multi-row headers, from images and PDFs into clean Markdown.

Document Processing
ocrtable-extractiondocument-processing
Browse all skills →